XGRIDS introduced PortalCam at NAB 2026, a spatial camera that combines LIDAR with a four-camera array to capture real-world environments as high-fidelity 3D assets. The device targets the gap between photogrammetry rigs that require large teams and smartphone 3D scans that lack detail.
Key takeaways:
LIDAR captures precise depth data while four cameras simultaneously capture color and texture from multiple angles
Outputs Gaussian Splatting files and detailed 3D reconstructions without complex setups or large teams
Designed for object reconstruction and environment capture in a single handheld device
What PortalCam Actually Does
PortalCam captures spatial data of real-world objects and environments and converts it into 3D digital assets. The hardware combines a LIDAR sensor for depth measurement with four cameras arranged to capture multiple angles simultaneously.
LIDAR provides the geometric skeleton. It measures distances with enough precision to build an accurate point cloud of whatever the device is pointed at. The four-camera array captures color, texture, and visual detail from different perspectives at the same time. The combination means the device captures both the shape and the appearance of objects in a single pass.
Gaussian Splatting Output
The captured data processes into Gaussian Splatting files, a 3D representation format that has gained traction in the computer graphics community for its balance of visual quality and rendering performance. Unlike traditional mesh-based models that require manual cleanup, Gaussian Splatting produces results that maintain detail from the original capture.
The output works in standard 3D workflows. Assets can be imported into game engines, visualization software, and virtual production environments. The format handles complex geometries like organic shapes, vegetation, and reflective surfaces that traditionally challenge photogrammetry.
Object and Environment Capture
PortalCam handles two capture modes. For object reconstruction, users scan individual items like props, products, or architectural elements. The device captures enough data to produce a detailed 3D model suitable for digital doubles, product visualization, or asset libraries.
For environment capture, users scan larger spaces like rooms, sets, or outdoor locations. The resulting 3D assets can serve as virtual production backgrounds, architectural references, or digital twins of physical spaces.
No Large Team Required
XGRIDS positioned PortalCam as lowering the barrier to high-fidelity 3D capture. Traditional photogrammetry rigs require multiple cameras mounted on rigs, controlled lighting, and often a team of technicians. PortalCam consolidates that into a single handheld device.
The tradeoff is capture volume. PortalCam works best for objects and environments that can be scanned from a reasonable distance. Large-scale terrain mapping or stadium-sized environments would still require drone-based or rig-based solutions.
XGRIDS PortalCam combines LIDAR depth sensing with a four-camera array to produce Gaussian Splatting 3D assets from a single handheld device.
